The pervasive issue of fraudulent transactions presents a considerable challenge for financial institutions globally. Developing innovative fraud detection systems is critical to maintaining customer confidence. Howev...
详细信息
ISBN:
(纸本)9783031752001;9783031752018
The pervasive issue of fraudulent transactions presents a considerable challenge for financial institutions globally. Developing innovative fraud detection systems is critical to maintaining customer confidence. However, several factors complicate the creating of effective and efficient fraud detection systems. Notably, fraudulent transactions are infrequent, resulting in imbalanced transaction datasets where legitimate transactions vastly outnumber instances of fraud. This data imbalance can concede the performance of fraud detection. Additionally, stringent data privacy regulations prevent the sharing of customer data, hindering the development of high-performing centralized models. Furthermore, fraud detection mechanisms must remain transparent to avoid impairing the user experience. This research proposes an approach utilizing Federated Learning (FL) with Explainable Artificial Intelligence (XAI) to overcome these obstacles. FL allows financial organizations to train fraud detection models collaboratively without requiring direct data sharing. So, customer confidentiality and data privacy are never compromised. Simultaneously, the incorporation of XAI guarantees that the model's predictions are interpretable by human experts. Experimental evaluations using real-time transaction datasets consistently demonstrate that the FL-based fraud detection system performs well. This study establishes the potential of FL as a reliable, privacy-preserving tool in combating fraud.
Modern computer graphics systems usually manage a set of internal data to store control parameters and all kind of graphics data. The behind problem on these graphics system data is that they are frequently updated an...
详细信息
Intelligent technologies are driving the development of smart campuses, fostering a dynamic and diverse intelligent ***, the current trend of customizing smart campus solutions often positions campus citizens as mere ...
详细信息
This study presents a novel translational method for suicide prevention research, employing customizable virtual reality (VR) simulations that mimic real-life situations. We aim to validate the usability of these VR s...
详细信息
Deep Neural Networks (DNNs) have found successful applications in various non-safety-critical domains. However, given the inherent lack of interpretability in DNNs, ensuring their prediction accuracy through robustnes...
详细信息
ISBN:
(纸本)9783031664557;9783031664564
Deep Neural Networks (DNNs) have found successful applications in various non-safety-critical domains. However, given the inherent lack of interpretability in DNNs, ensuring their prediction accuracy through robustness verification becomes imperative before deploying them in safety-critical applications. Neural Network Verification (NNV) approaches can broadly be categorized into exact and approximate solutions. Exact solutions are complete but time-consuming, making them unsuitable for large network architectures. In contrast, approximate solutions, aided by abstraction techniques, can handle larger networks, although they may be incomplete. This paper introduces AccMILP, an approach that leverages abstraction to transform NNV problems into Mixed Integer Linear Programming (MILP) problems. AccMILP considers the impact of individual neurons on target labels in DNNs and combines various relaxation methods to reduce the size of NNV models while ensuring verification accuracy. The experimental results indicate that AccMILP can reduce the size of the verification model by approximately 30% and decrease the solution time by at least 80% while maintaining performance equal to or greater than 60% of MIPVerify. In other words, AccMILP is well-suited for the verification of large-scale DNNs.
The integration of artificial intelligence in agriculture has revolutionized farming practices, enhancing crop yields and resource efficiency. However, existing machine learning systems primarily focus on livestock, o...
详细信息
In the field of software development, ensuring the accuracy and quality of code remains a paramount concern. The task of precisely classifying code as correct or incorrect poses inherent challenges. This research intr...
详细信息
Testing complex systems is crucial for ensuring safety, especially in automated driving, where diverse data sources and variable environments pose challenges. Here, robust safety validation is critical but exhaustive ...
详细信息
Reactive synthesis is an automated process for deriving correct-by-construction reactive systems from temporal specifications. GR(1), in particular, is a popular LTL fragment that balances efficient synthesis complexi...
This paper examines the blessings of dynamic communications software program design in networks. By leveraging advances in generation, dynamic communications provide an ever-evolving method of connecting, communicatin...
详细信息
暂无评论